### LLaMA-Factory SFT Training Config ###
### For session-level reflection training ###

### Model
model_name_or_path: /projects/bfqt/users/yurenh2/ml-projects/personalization-user-model/models/llama-3.1-8b-instruct

### Method
stage: sft
do_train: true
finetuning_type: full
deepspeed: ds_z3_config.json

### Dataset
dataset: sft_reflection
dataset_dir: /projects/bfqt/users/yurenh2/ml-projects/personalization-user-model/collaborativeagents/training
template: llama3
cutoff_len: 4096

### Output
output_dir: /projects/bfqt/users/yurenh2/ml-projects/personalization-user-model/collaborativeagents/training/outputs/sft_reflection_lf

### Training
per_device_train_batch_size: 1
gradient_accumulation_steps: 16
learning_rate: 1.0e-6
num_train_epochs: 4.0
lr_scheduler_type: cosine
warmup_ratio: 0.1
bf16: true
ddp_timeout: 3600

### Logging
logging_steps: 10
save_steps: 100
save_total_limit: 3